Deep Research: A Survey of Autonomous Research Agents

2025-08-18

AI Summary

Large language models (LLMs) suffer from static knowledge bases, limiting their ability to generate credible, evidence-based analytical reports grounded in real-time web data. To address this, we propose the “Deep Research Agent” paradigm—a four-stage technical pipeline encompassing question planning, adaptive querying, goal-directed web exploration, and evidence-driven report generation. We systematically formalize key technical challenges across all stages and introduce a synergistic optimization framework integrating active retrieval, multi-step reasoning, task decomposition, and information provenance tracking. Furthermore, we construct a dedicated benchmark for rigorous evaluation. Experiments demonstrate that our agent achieves end-to-end, traceable, and high-fidelity autonomous research on complex questions, significantly improving answer accuracy and evidence coverage. This work establishes both a theoretical foundation and a practical methodology for building trustworthy intelligent agents capable of dynamic knowledge expansion.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Modeling other Agents
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

The rapid advancement of large language models (LLMs) has driven the development of agentic systems capable of autonomously performing complex tasks. Despite their impressive capabilities, LLMs remain constrained by their internal knowledge boundaries. To overcome these limitations, the paradigm of deep research has been proposed, wherein agents actively engage in planning, retrieval, and synthesis to generate comprehensive and faithful analytical reports grounded in web-based evidence. In this survey, we provide a systematic overview of the deep research pipeline, which comprises four core stages: planning, question developing, web exploration, and report generation. For each stage, we analyze the key technical challenges and categorize representative methods developed to address them. Furthermore, we summarize recent advances in optimization techniques and benchmarks tailored for deep research. Finally, we discuss open challenges and promising research directions, aiming to chart a roadmap toward building more capable and trustworthy deep research agents.

Problem

• Overcoming LLM knowledge boundaries via autonomous research agents
• Surveying deep research pipeline stages and technical challenges
• Advancing trustworthy autonomous research agent development

Innovation

• Autonomous agents perform web-based evidence retrieval
• Pipeline includes planning, question, exploration, reporting
• Optimization techniques enhance research agent capabilities